| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Biochemical Sciences ‘A. Rossi Fanelli’, University of Rome La Sapienza, Rome, Italy;
C.N.R. Institute of Biophysics, Pisa, Italy;
Department of Experimental Medicine and Pathology, University of Rome La Sapienza, Rome, Italy; and
§ DIBIT, Department of Biological and Technological Research-DIBIT, S. Raffaele Scientific Institute, Milano, Italy
1Correspondence: Department of Biochemical Sciences, University of Rome ‘La Sapienza’, P.le Aldo Moro 5, 00185 Rome, Italy. E-mail: rodolfo.ippoliti{at}uniroma1.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: plasminogen activator • ribosome-inactivating proteins • receptors • ligand-passing • cancer
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, melanomas (2)
, colon (3)
,
and prostate carcinomas (4)
. uPAR is a receptor that
preferentially binds species-specific Pro-uPA/uPA or its derived
amino-terminal fragment (ATF) (5)
. However, ATF and the
secreted precursor Pro-uPA (p-uPA) are not internalized, whereas
uPAR-bound uPA internalizes only if complexed to plasminogen activator
inhibitors (PAIs). Endocytic receptors such as the 
2-macroglobulin
receptor or LDL-related receptor protein (LRP), the very low density
lipoprotein receptor, and the epithelial glycoprotein-330, are able to
mediate the internalization of uPAR-bound uPA/PAI-1 complexes (6
, 7)
. Both uPAR and LRP are involved in the endocytosis
(8)
; after ligand release, they recycle back to the plasma
membrane (9)
.
Saporin belongs to a class of monomeric plant seed
ribosome-inactivating proteins (type I RIPs, 10
) that
catalyzes the in vitro depurination of a specific adenine
residue in large ribosomal RNAs (11)
. Lacking a membrane
binding subunit, such as in type II RIPs (i.e., ricin), they usually
cannot enter cells by receptor-mediated endocytosis unless complexed to
an appropriate carrier molecule; nevertheless, free saporin (and
presumably other type I RIPs) can be internalized in some cells
expressing LRP receptors (12)
, thereby acquiring
cytotoxicity. A conjugate between active two-chain human uPA and native
saporin was shown to internalize in the absence of plasminogen
activator inhibitors (13)
in cells expressing both uPAR
and LRP receptors. Moreover, preparation of an ATF-saporin recombinant
chimera demonstrated that human ATF domain directs this chimera to
human uPAR-bearing cells and that the toxin domain can mediate
internalization (14)
. In fact, competition experiments
showed that the saporin moiety is involved in the internalization of
ATF-saporin through LRP receptors. Thus, neither catalytically active
urokinase nor PAIs are needed to initiate these internalization
pathways.
In this paper we focus on the intracellular pathway of a related
chimeric toxin (p-uPA-SapTRITC), between
fluorescently labeled saporin and human pro-urokinase in murine cells
transfected (15)
with the human uPAR gene (Cl19). Again,
this conjugate was found to be targeted to receptors (uPAR and LRP)
that do not colocalize in control cells. uPAR is in fact, as previously
shown, mostly confined to the leading edge of migrating cells
(16
, 17)
, whereas LRP is distributed among the whole
plasma membrane (18)
and inside clathrin-coated
invaginations. Fluorescence microscopy experiments show that binding
to the cell membrane is indeed mediated by uPAR, but on triggering
of the endocytic process, p-uPA-SapTRITC seems to
follow the fate of internalized LRP.
In contrast to what is proposed for the endocytosis of uPA-PAI complexes, where uPAR and LRP receptors remain part of the same quaternary complex (presumably through the presence of PAI), we propose that endocytosis of p-uPA-SapTRITC follows a two-step process. Initially the p-uPA domain of the toxin binds to uPAR, presumably triggering a signal that induces migration of LRP toward uPAR; by virtue of some conformational change, this is subsequently coupled to an exchange of the chimeric toxin (mediated by the saporin domain) between the two receptors such as p-uPA-SapTRITC follows LRP in the endocytic process.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Synthesis of the conjugate
Saporin was purified and labeled with rhodamine as described
(19)
. The rhodaminated toxin was conjugated with p-uPA
following the procedure of Cavallaro et al. (13)
with some
modifications. SapTRITC was modified with 2.5-fold molar
excess of 2-iminothiolane (instead of SPDP) to introduce a free
sulfydril group and allow the following reaction with SPDP-modified
p-uPA.
The p-uPA-SapTRITC conjugate was purified by HPLC
on a ion exchange column (Vydac VHP575) as described (13)
.
In vitro ribosome-inhibiting activity
The RIP activities of rhodaminated saporin and
p-uPA-SapTRITC were measured in
nuclease-treated rabbit reticulocyte lysates (Promega, Madison, Wis.)
measuring BMV mRNA translation as described (20)
. Values
were expressed as percent of control 3H-leucine
incorporation by the untreated lysates. Activity was calculated by
measuring the concentration of toxin inhibiting mRNA translation by
50% (IC50).
Cell-killing experiments
At least two independent experiments were performed, using
murine fibroblast LB6 and Cl19 expressing human uPAR treated as
described (13)
. Cytotoxicity was calculated by measuring
the dose of toxin inhibiting by 50% the incorporation of the untreated
cells (ID50). For the competition experiments,
Cl19 cells were plated at 104 cells in 80 µl
and exposed to 5 x 10-9 M
p-uPA-SapTRITC conjugate in the absence or in the
presence of increasing concentrations of human recombinant pro-uPA
(from 25x10-9 M to
250x10-9 M), for a total exposure of 2 h.
Cells were acid-washed, incubated 16 h at 37°C, then
pulse-labeled with 3H-leucine for 4 h.
Protein synthesis was not affected by pro-uPA exposure. Each experiment
was performed in four replicates.
FACS scan analysis
Fluorescence-activated cell sorting was performed essentially as
described (8)
using aliquots of 106
U937 cells exposed 90 min at 4°C either to 10 nM p-uPA alone and a
preformed uPA-PAI1 complex as control or to 10 nM p-uPA-Sap TRITC.
Briefly, after exposure to the ligands, the U937 cells were washed
extensively and duplicate samples warmed up to 37°C for 0, 5, 10, 15,
30, 45, 60, and 90 min. After transfer to an ice bath to stop the
internalization process, the samples were washed and further incubated
at 4°C with R2 MoAb (30 µg/ml) or anti-LRP polyclonal antibodies,
followed by incubation with FITC anti-mouse or FITC anti-rabbit
secondary antibodies, respectively. Background fluorescence was
estimated with samples of cells exposed to the ligand (s) and incubated
only with the secondary antibodies. Fluorescence of 0.5–1 x
104 cells per sample was usually read in
duplicate samples.
Fluorescence microscopy
Topology of uPAR and LRP in control and
p-uPA-SapTRITC-treated cells
Cl19 cells grown on coverslips were incubated with nothing or
p-uPA-SapTRITC at a 300 nM concentration in D-MEM
(Life Technologies, Inc.) containing geneticin sulfate G418 for 60 min
at 4°C. The cells were then washed with cold medium and further
incubated at 37°C in the humidified chamber (in the presence of 5%
CO2) for different times. The cells were then
fixed with 4% p-formaldehyde in phosphate-buffered saline (PBS) and
permeabilized with 0.1% Triton X-100. Immunochemical staining of the
two receptors was done by incubation of the cells with the monoclonal
anti-uPAR (R2, 0.125 mg/ml) or the rabbit
polyclonal anti-LRP (0.05 mg/ml), followed by secondary antibodies
conjugated with fluorescein or rhodamine.
Coverslips were then mounted on slides with Aqua polymount (Polyscience Inc., Warrington, Pa.) and observed with either a Zeiss-Axiophot microscope connected to a Hamamatsu Argus 20 video recording system or a confocal microscope (custom assembled by P. A. Benedetti).
Endocytosis of p-uPA-SapTRITC in the presence of
R2 MoAb anti-uPAR
An experiment in the presence of the R2
MoAb anti-uPAR was carried out incubating cells grown on coverslips
with p-uPA-SapTRITC in the presence of the
R2 MoAb anti-uPAR (125 µg/ml) at 4°C for 60
min. After washing, cells were immediately fixed or transferred to
37°C for further incubation and then fixed. uPAR was visualized by
addition of a secondary anti-mouse antibody FITC-labeled.
Endocytosis of p-uPA-SapTRITC in the presence of
filipin or chlorpromazine
Cells grown on coverslips were incubated when indicated with 1
µg/ml Filipin III (Sigma) or with 10 µg/ml chlorpromazine (Sigma)
in the culture medium for 30 min at 37°C. The cells were incubated
with 300 nM p-uPA-SapTRITC for another 30 min,
washed, fixed, and observed under the microscope. Where indicated, a
fluorescent derivative of the toxin lectin Ricin
(RCAFITC, 300 nM) was used for the second
incubation.
Endocytosis of p-uPA-SapTRITC followed in Cl19 living
cells
In some experiments cells were incubated with
p-uPA-SapTRITC as described above and further
incubated with NBD-ceramide (21)
for visualization of the
Golgi apparatus in living cells.
Immunoelectron microscopy
Cl19 cells were treated with
p-uPA-SapTRITC 300 nM for 1 h at 4°C and
immediately fixed or warmed to 37°C for 5 min to allow
internalization. Cl19 untreated cells were used as a control. All cells
were washed twice with PBS and fixed in 4% paraformaldehyde for 30 min
at 25°C. For single labeling experiments, cells were scraped and
incubated with anti-uPAR monoclonal antibody (0.2 mg/ml in PBS 1%
bovine serum albumin (BSA) for 1 h at 4°C) or with anti-LRP
polyclonal antibody (0.5 mg/ml in PBS 1% BSA for 1 h at 4°C).
Cells were then washed extensively and labeled for 3 h at 4°C
with colloidal gold (18 nm, prepared by citrate method) conjugated with
protein A (Pharmacia/Upjohn). For double labeling experiments, cells
were scraped and incubated with anti-uPAR monoclonal antibody (0.2
mg/ml in PBS 1% BSA for 1 h at 4°C), followed by 10 nm goat
anti-mouse immunoglobulin G gold conjugates and then with anti-LRP
polyclonal antibody (0.5 mg/ml in PBS 1% BSA for 1 h at 4°C),
followed by 18 nm protein A gold particles. Samples were postfixed in
1% osmium tetroxide in Veronal acetate buffer, pH 7.4, for 2 h at
4°C, stained with uranyl acetate (5 mg/ml), dehydrated in acetone,
and embedded in Epon 812.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
|
In cell killing experiments, the conjugate was extremely cytotoxic for
murine-transfected Cl19 cells expressing uPAR, with an
ID50 value of 0.015 nM; free saporin showed an
ID50 of 100 nM on the same cells, as previously
shown (20)
.
When tested on the untransfected wild-type cells (LB6),
p-uPA-SapTRITC had an ID50
of 2 nM (133-fold higher than in Cl19 cells) while Sap toxicity, as
expected, remained unchanged (ID50=100 nM, Table 1
). Furthermore, p-uPA-SapTRITC toxicity (over
three orders of magnitude higher in potency than free saporin) may be
competed by addition of an excess of free human p-uPA to Cl19 cells
(data not shown).
The topology of uPAR and LRP in LB6 and Cl19 cells
Immunocytochemical staining on wild-type LB6 showed that
nontransfected LB6 cells do not express human uPAR, and indeed no
staining was observed either on the cell surface (not permeabilized
cells) or intracellularly (with Triton X-100 permeabilized cells).
Instead, LRP was found throughout the membrane of both LB6 and Cl19
cells and heavily stained the perinuclear (ER) region in permeabilized
cells (data not shown).
In Cl19 cells, human uPAR is overexpressed and appears to be localized
in permeabilized cells in the perinuclear region, where it is
colocalized with LRP (presumably as newly synthesized polypeptides). On
the plasma membrane of nonpermeabilized cells, however, the two
receptors are not colocalized, uPAR being mostly confined to the focal
adhesion regions of the cell whereas LRP stained uniformly the membrane
and scarcely overlapped with uPAR (see Fig. 1
).
|
p-uPA-SapTRITC binding and internalization in Cl19
cells
p-uPA-SapTRITC binds to Cl19
cells and is rapidly internalized at 37°C (Fig. 2a
). Although saporinTRITC
is toxic to both LB6 and Cl19 cells (see Table 1
), we could not reveal
its presence in fluorescent spots as with
p-uPA-SapTRITC-treated cells (see Fig. 2b
). This difference can be attributed to the higher
efficiency of the binding and internalization mediated by the presence
of the pro-urokinase domain conjugated to saporin. The fluorescence
signal due to the saporin molecule is quite clear (Fig. 2a
)
and cannot be confused with that due to autofluorescence (Fig. 2c
).
|
As shown by confocal microscopy, after binding at 4°C (see Fig. 3
) the chimeric toxin (red fluorescence) was mostly colocalized (Fig. 3a
) with uPAR (green fluorescence) but scarcely (if at all;
Fig. 3b
) with LRP (green fluorescence).
|
When the temperature was raised to 37°C the chimera was internalized,
and after 30 min (Fig. 3d
) appeared in clearly visible spots
(presumably endosomes) that largely overlapped with the fluorescence
signal due to LRP localization. On the contrary, uPAR still seemed
mostly confined to the edges of the membrane and did not appear to
enter the cell after endocytosis (Fig. 3c
). Even when
shorter times of incubation at 37°C were used we could not observe
colocalization between uPAR and p-uPA-SapTRITC.
The conclusion of these observations leads us to hypothesize a
different possible role of the two receptors, with uPAR being essential
for binding to the cell membrane and LRP for internalization.
To test this hypothesis, we followed binding and internalization of
p-uPA-SapTRITC in the presence of a MoAb
anti-uPAR (R2). This antibody does not compete
for the ligand binding site of p-uPA and has no effect on the endocytic
process as it has been widely used to follow the down-regulation of
uPAR (8
, 9)
. As shown in Fig. 4
, the R2Ab signal (green) overlaps that of the
toxin (red) at 4°C (Fig. 4a
), but when
p-uPA-SapTRITC is internalized as a function of
time at 37°C (Fig. 4b
, c
, d
), the green signal remains mainly
confined to the edges of the membrane. The presence of the antibody
does not induce any change on the distribution nor does it trigger
endocytosis of the receptor (data not shown).
|
We followed as control the cellular localization of uPAR and LRP on
binding and internalization of the complex uPA-PAI1, known conversely
to induce uPAR endocytosis and recycling (8
, 9)
. Indeed,
the two receptors (red uPAR, green LRP) still appeared segregated into
different regions of the cell membrane after binding of uPA-PAI1
complex at 4°C (Fig. 4e
), but showed a distinct
colocalization after 30 min at 37°C (Fig. 4f
).
Moreover, further evidences were obtained by cytofluorometric analysis
of U937 human monocytes exposed to p-uPA-SapTRITC
or the uPA-PAI1 complex, as a control. Down-regulation of uPAR after
binding and internalization of the uPA-PAI1 complex was indeed observed
as previously reported (9)
, whereas neither p-uPA alone
nor p-uPA-SapTRITC could induce any change in the
levels of uPAR on the cell membrane even at the long incubation times
(data not shown).
Fluorescence microscopy in the presence of filipin and
chlorpromazine
To further demonstrate that the internalization of
p-uPA-SapTRITC is mediated by endocytosis through
coated pits and not by caveolae, the following experiments have been
carried out in the presence of filipin and chlorpromazine. Filipin is a
drug that interferes with the normal distribution of cholesterol by its
sequestration. This process induces a substantial loss of
‘caveolae-like’ structures blocking transcellular transport in
endothelial cells (22)
and blocks the intracellular
absorption of cholera toxin (23)
, a protein known to
interact with ganglioside GM1 in the membrane and to follow the
‘clathrin-independent’ endocytosis pathway. Since uPAR is a
GPI-anchored receptor localized in caveolae in some cells
(24)
, we tested whether the presence of filipin could
induce changes in the normal distribution of the receptor inside the
membrane and whether the endocytic process of
p-uPA-SapTRITC conjugate in CL19 cells could be
specifically inhibited.
Filipin did not induce any significant redistribution of either uPAR or
LRP after 30' incubation at 37°C (data not shown); furthermore, the
drug did not inhibit the binding and internalization of
p-uPA-SapTRITC conjugate, as shown in Fig. 5c
(compare with Fig. 5a
, as a control).
|
Chlorpromazine is a known inhibitor of clathrin coat assembly and
completely blocks receptor-mediated endocytosis of many molecules
(23
, 25)
. In our experimental conditions, this drug could
block internalization (but not binding) of
p-uPA-SapTRITC in CL19 cells (see Fig. 5e
) as visualized by the appearance of patches on the cell
membrane that contained the conjugate still bound to uPAR; indeed, uPAR
distribution also changed after exposure to chlorpromazine at 37°C,
with the appearance of patches (not shown). Surprisingly, LRP was not
affected by the presence of chlorpromazine in the absence of ligand.
As a control we used fluorescently labeled ricin to verify the activity
of chlorpromazine and filipin. As expected, most of the toxin remains
on the plasma membrane in the presence of chlorpromazine (see Fig. 5f
), whereas filipin seemed to have scarce effect (if any)
on the endocytosis of the toxin (see Fig. 5d
) that
accumulated in the Golgi apparatus as in the control cells (Fig. 5b
).
Immunoelectron microscopy
Immunogold electron microscopy was used to assess the relative
position of uPAR and LRP on the membrane of Cl19 cells on binding and
internalization of p-uPA-SapTRITC. In both
untreated (Fig. 6a
) or treated with
p-uPA-SapTRITC at 4°C (Fig. 6b, c
)
Cl19 cells, the two immunogold-labeled receptors appeared to be
localized on the surface microvilli (Fig. 6
, arrows) and they partly colocalized (large golds refer to LRP, small
golds to uPAR); in addition, the receptors appeared to be excluded from
clathrin-coated pits as shown in single labeling for uPAR (Fig. 6b
, arrowhead).
|
During 5' at 37°C internalization after the binding of the conjugate,
we observed a colocalization of the two receptors in close proximity to
clathrin-coated pits (Fig. 6d, e
, arrowheads), at the base
of microvilli (Fig. 6e
, arrows) and inside endocytic
invaginations (Fig. 6f
, arrowhead). Thus, at least at the
very beginning of the endocytic process, uPAR and LRP are moving
together toward that region of the membrane where the endocytic pits
are forming.
In vivo localization of p-uPA-SapTRITC
As shown in Fig. 7
, in living (not fixed) cells the toxic
p-uPA-SapTRITC was accumulated in vesicular
structures that did not overlap the fluorescence of the Golgi marker
NBD-ceramide (21)
; the fluorescent spots varied in
dimension and positions, starting from the plasma membrane down to the
TGN, allowing one to hypothesize a bulk transport along the endosomal
system to a possible final degradation in the lysosomes. No direct
evidences were available at this resolution as to a possible escape of
free saporin or the intact conjugate from the endocytic compartment(s)
to the cell cytoplasm.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and recently a strategy has been devised allowing to
express a secretory version of a recombinant chimera between the
amino-terminal fragment of human urokinase and saporin (ATF-Sap) in
high amounts in eukaryotic cells (27)
. The possibility of
investigating the intracellular pathways, followed by these chimeras
represents a prerequisite for optimizing and modulating their
therapeutic use. To dissect and understand the sequence of steps
involved in binding and internalization and to study the transport
along the endomembrane system, we used the model cell line of murine
cells Cl19 transfected with human uPAR, for which data are available
(3
, 15)
on uPAr and LRP levels (1.2x10
5 and 1–5x10 4 per
cell respectively) and followed cell-surface binding and
internalization of p-uPA-SapTRITC conjugate. This
chimera represents a unique tool to study the interactions between uPAR
and LRP receptors and to obtain insights on intracellular routing of
single chain RIPs such as saporin. The presence of uPAR confers to
p-uPA-SapTRITC an extreme toxicity toward cells
overexpressing this receptor, thus making the system very sensitive.
The receptor of urokinase-type plasminogen activator is an
adhesion/migration receptor that focuses the urokinase activity at the
leading edge of actively migrating cell (28)
; it behaves
as an activable cell-surface chemokine capable of triggering signaling
in a variety of cell types (29
30
31)
; moreover, it has
recently been demonstrated that cooperation between uPA/uPAR and
metalloproteinase MMP-9 is required for metastasis spread
(32)
. uPAR and the endocytic receptor LRP normally do not
colocalize, and indeed some signaling or ‘cross-talk’ between these
two receptors might exist: ATF has been shown to induce Cl19 cell
migration, and these phenomena are mediated by a signal transduction
cascade (33
, 34)
. Moreover, ATF-Sap is efficiently
endocytosed (14)
, and down-regulation of LRP by FACS-scan
analysis after binding and internalization of the recombinant chimera
to U-937 cells could be observed (M. S. Fabbrini, unpublished
results), indicating that LRP is actively involved in the endocytosis
of ATF-Sap (and presumably of p-uPA-SapTRITC).
Physiological regulation of uPA activity on the cell membrane is
mediated by internalization of inhibitor complexes, such as uPA-PAI1,
with a mechanism involving cointernalization of both uPAR and LRP
receptors (9)
; the saporin domain X-linked to p-uPA is
expected to mimic the PAI1 function in anchoring the complex to LRP,
and thus one would expect that p-uPA-SapTRITC
followed an endocytic pathway similar to uPA-PAI1. However, both
fluorescence microscopy on Cl19 and experiments by FACS scan analysis
of U937 cells after exposure to p-uPA-SapTRITC
indicated no down-regulation of uPAR. Altogether, our results seem to
indicate that only LRP is responsible for the internalization of
p-uPA-SapTRITC.
Whereas the intracellular trafficking of chimeric toxins is generally
dictated by the ligand moieties (19)
, in this case
internalization seems to take place just because of the presence of the
toxin moiety, which binds to the endocytic receptor LRP. uPAR binding
to the uPA moiety allows p-uPA-SapTRITC to fix on
the plasma membrane, and the subsequent transfer of the complex to LRP
may require a possible cross-signaling between the two receptors. These
results are indeed supported by data previously available
(35)
on the relative affinities of p-uPA for uPAr and LRP
(5x10-11 and 1x10 -8 M,
respectively) and by the observation that the presence of uPAr
completely abolishes the interaction of p-uPA-PAI1 complex with LRP.
Direct observation of p-uPA-SapTRITC in single cells allowed us to follow all the phases involved in the endocytic process, starting from the binding to the plasma membrane. As clearly shown by confocal microscopy, the toxic conjugate binds only to uPAR at 4°C, since binding does not overlap with LRP distribution. By increasing the temperature to 37°C, endocytosis starts and the conjugate is internalized, but inside the cell it is mostly observed to coincide with LRP, whereas uPAR is still confined at the leading edges of the cell membrane. The endocytosis of p-uPA-SapTRITC conjugate presumably proceeds following the classical pathway mediated by the clathrin coat formation, as demonstrated by the inhibition induced by chlorpromazine. Our data might indicate an interaction of uPAR with clathrin components prior to ligand binding; on the contrary, filipin alters the structure of the caveolae but does not inhibit p-uPA-SapTRITC internalization. Altogether, these results favor LRP as the main receptor involved in the endocytic process.
More evidence that uPAR may not be involved in the intracellular pathway, followed by p-uPA-SapTRITC, is provided by the experiment in which living Cl19 cells were incubated with both the conjugate and the monoclonal anti-uPAR at 4°C. In this condition the conjugate was internalized after increasing the temperature to 37°C, but the antibody (and hence the receptor) remained on the cell surface.
Immunoelectron microscopy revealed that after exposure at 4°C to p-uPA-SapTRITC, the two receptors appear to be excluded from endocytic invaginations. Once the cells exposed to p-uPA-SapTRITC have been warmed to 37°C for 5 min, both LRP and uPAR move toward the endocytic-forming pits and are present in unclosed invaginations, which may be defined as coated pits.
We have never observed by fluorescence microscopy any clear
colocalization of uPAR and p-uPA-SapTRITC during
internalization; however, the higher resolution of electron microscopy
allows us to better discriminate the earliest phases of this process. A
possible recycling of uPAR just below the plasma membrane on triggering
of endocytosis might be suggested, being the
p-uPA-SapTRITC molecule presumably transferred to
LRP inside the forming endocytic vesicle. LRP might then recycle to the
plasma membrane whereas p-uPA-SapTRITC is
presumably delivered via the endosomal system to lysosomes, where
degradation might take place. Indeed, bafilomycin A and chloroquine
treatment of intoxicated cells potentiated ATF-saporin cytotoxicity
(14)
, suggesting a possible involvement of acidic
compartments in the routing of the toxin. Part (not detectable) of the
conjugated or free saporin must escape the vesicles and reach the
ribosomes in a pathway not yet clarified. It must be stressed that the
bulk of p-uPA-SapTRITC does not to seem to reach
the Golgi complex. Also, recombinant ATF-Sap to which a KDEL sequence
has been added was no more cytotoxic than the parental ATF-Sap (M. S. Fabbrini et al., unpublished results) in U937 cells, suggesting that
the Golgi complex indeed might not be involved in toxin routing, in
contrast to what is known in the case of ricin (36)
.
Based on the summary reported above, our hypothesis about the endocytic
process of the toxic p-uPA-SapTRITC conjugate may
be depicted as in Fig. 8
.
|
This mechanism of binding to the plasma membrane might resemble that
described for tumor necrosis factor (TNF), where a dual step process
involving two receptors for the TNF has been identified and called
‘ligand passing’ (37)
: TNF binds on the plasma membrane
through the first receptor (P55); once transfer has occurred, the
second receptor (P75) triggers the intracellular second messenger
cascade.
In this work we describe a somewhat similar mechanism, although the ‘ligand passing’ event involves two different receptors: uPAR, which recognizes the pro-urokinase moiety warranting a correct anchorage and orientation of p-uPA-SapTRITC on the plasma membrane; and LRP, which binds saporin and actively transports the conjugate inside the cell.
In conclusion, we have described a peculiar and novel mechanism of
endocytosis of a toxic conjugate differing from that of the ligand used
for cellular targeting. Since many of the conjugated (immuno- or
hormono-) toxins used in cancer therapy approaches contain saporin (or
similar type I RIPs; ref 38
), it is important to assess
whether their targeting and toxicity may be influenced by the presence
of cooperative effects between receptors. In fact, although most
research in this field has now converged to a careful molecular design
to obtain recombinant toxins (39)
with a higher degree of
specificity and toxicity, scarce attention has been devoted to the
study of molecular interactions at the level of the plasma
membrane.
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 25, 1999.
Revision received December 16, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-2 macroglobulin receptor/LDL receptor-related protein (Lrp)-dependent internalization of the urokinase receptor. J. Cell Biol. 131,1609-1622-2 macroglobulin receptor mediates binding and cytotoxicity of plant ribosome-inactivating proteins. J. Biochem. 232,165-171
-2 macroglobulin receptor and epithelial glycoprotein-330: two giant receptors mediating endocytosis of multiple ligands. Biochim. Biophys. Acta 1197,197-213[Medline]
2-macroglobulin receptor/low density lipoprotein receptor-related protein. J. Biol. Chem. 269,25668-25676This article has been cited by other articles:
|
|
 |
|
  G. Zhang, H. Kim, X. Cai, J. M. Lopez-Guisa, P. Carmeliet, and A. A. Eddy Urokinase Receptor Modulates Cellular and Angiogenic Responses in Obstructive Nephropathy J. Am. Soc. Nephrol., May 1, 2003; 14(5): 1234 - 1253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Liu, T. H. Bugge, and S. H. Leppla Targeting of Tumor Cells by Cell Surface Urokinase Plasminogen Activator-dependent Anthrax Toxin J. Biol. Chem., May 18, 2001; 276(21): 17976 - 17984. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
